MRI: Acquisition of a Heterogeneous High-Performance Computing Cluster Driven by Computational and Data-Intensive Multidisciplinary Research

MRI：获取由计算和数据密集型多学科研究驱动的异构高性能计算集群

• 批准号: 2216311
• 负责人: Kevin Brandt
• 金额: $ 97.87万
• 依托单位: South Dakota State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-10-01 至 2025-09-30
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2216311&HistoricalAwards=false
• 关键词: MRI; Acquisition; Heterogeneous; Performance; Computing

South Dakota State University (SDSU) will enable regional research and educational programs with high computational and data-intensive needs through the acquisition and deployment of an advanced computing instrument. The goals of this initiative will accelerate scientific discovery, promote active engagement, advance collaboration within the research community, and provide a resource within the region to advance and elevate multi-disciplinary STEM education and research. A breadth of science and engineering fields will benefit from this acquisition, including climatology, satellite image processing, national food safety and security, genomics of bacterial infections, antibiotic resistance, and power grid security. This project will also be a focal point for supporting the educational and research needs of underserved communities throughout the region. The team within SDSU Research Cyberinfrastructure (RCi) will acquire and implement a robust advanced computing instrument that accomplishes the following key project objectives: (1) Widening and boosting computing core capabilities within research pipelines that harness the power of vector and parallel computing; (2) Advancing the usability of CPU and Graphics Processing Unit (GPU) resources using Open OnDemand and Science Gateways; and (3) Amplifying the accessibility of highly cost-effective computational tools to a larger user base. The system design uses a next-generation computer architecture with hardware resources specifically designed for boosting computing capacity. The instrument will offer the flexibility of scaling research applications, allowing the user to self-provision resources and facilitating ease of access for new researchers. This will be achieved using off-the-shelf open-source software resources and pre-built scientific workflows. This instrument will serve as a computational core for RCi at SDSU, serving researchers, collaborators, staff, and students engaged in research and education.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

南达科他州州立大学（SDSU）将通过收购和部署先进的计算仪器，实现具有高计算和数据密集型需求的区域研究和教育计划。该倡议的目标将加速科学发现，促进积极参与，促进研究界内部的合作，并在该地区提供资源，以促进和提升多学科STEM教育和研究。广泛的科学和工程领域将受益于此次收购，包括气候学，卫星图像处理，国家食品安全和保障，细菌感染基因组学，抗生素耐药性和电网安全。该项目还将成为支持整个区域服务不足社区的教育和研究需求的协调中心。SDSU Research Cyberinfrastructure（RCi）内的团队将获得并实施一个强大的先进计算工具，以实现以下关键项目目标：（1）扩展和提升研究管道内的计算核心能力，利用向量和并行计算的力量;（2）使用开放式按需和科学网关提高CPU和图形处理单元（GPU）资源的可用性;及（3）扩大高成本效益计算工具的使用范围，让更多用户使用。系统设计采用下一代计算机架构，硬件资源专门用于提高计算能力。该工具将提供扩展研究应用的灵活性，允许用户自行提供资源，并方便新研究人员的访问。这将通过使用现成的开源软件资源和预先构建的科学工作流程来实现。该仪器将作为SDSU RCI的计算核心，为从事研究和教育的研究人员、合作者、工作人员和学生提供服务。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。